Otherindependent cross-checkabstract only checked I-UPF-02 standard review

How reliable is the NOVA classification?

Ultra-processed foods: how functional is the NOVA system?

Overview

Braesco V et al.. Eur J Clin Nutr 2022 · PMID 35314769 · DOI 10.1038/s41430-022-01099-1

Experts often disagreed when sorting foods into NOVA groups, though typical UPF were mostly classified alike.

Does the study support the specific claim? No · Methodological quality: moderate

Study explained simply

What did the researchers want to find out?: Does NOVA lead to consistent food assignment?
How was the study done?: Online survey of French food/nutrition specialists assigning foods to NOVA groups.
How many people, animals or samples were studied?: 159 (marketed foods) and 177 (generic foods) evaluators; 120 + 111 foods; French food and nutrition specialists
What was actually found?: Fleiss kappa 0.32 (marketed foods with ingredients) and 0.34 (generic foods); one cluster of 90 marketed foods mostly assigned NOVA 4 (91%); other clusters heterogeneous.
How large was the observed effect?: Kappa 0.32-0.34
What do the results mean in everyday life?: UPF category is useful but fuzzy at the edges.
What are the study's limitations?: French experts only; Authors with industry links
What does it say about Elawady's specific claim?: Contradicts C-UPF-01 (precision).
Do the results apply to Germany and Europe?: High.
What can explicitly not be concluded from this study?: That NOVA is useless.

Authors: Braesco V et al.
Journal: Eur J Clin Nutr
Year: 2022
Identifier: PMID 35314769 DOI 10.1038/s41430-022-01099-1
Verification: verified – PubMed efetch 2026-10-25
Retracted: no
Corrections: none
Extent checked: abstract only
Conflicts of interest and funding: Authors report grants/consultancy including food industry (declared)

Does NOVA lead to consistent food assignment?

Study design: Other – Online survey of French food/nutrition specialists assigning foods to NOVA groups.
Population: French food and nutrition specialists
Country: France
Period: not stated in abstract
Sample size: 159 (marketed foods) and 177 (generic foods) evaluators; 120 + 111 foods
Inclusion and exclusion criteria: not stated in abstract

Results: Fleiss kappa 0.32 (marketed foods with ingredients) and 0.34 (generic foods); one cluster of 90 marketed foods mostly assigned NOVA 4 (91%); other clusters heterogeneous.
Absolute risks: not applicable
Authors' conclusion: NOVA criteria do not allow robust and functional assignments.

Kappa ~0.3 = fair agreement.

Contradicts C-UPF-01 (precision).

Strong agreement for many typical UPF (cluster of 90 foods).

High.

Important methodological critique.

Multidimensional assessment
Multidimensional assessment
Authenticity of the source	verified
Methodological quality	moderate
Risk of bias	some concerns
Precision of results	moderate
Directness	direct
Reproducibility	not replicated
Consistency with the overall evidence	unknown
Regional transferability	high
Adequacy of the interpretation (by the archive)	not applicable

Reliability of revised criteria.